Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    instances of events or actions that can be done quickly

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to brief or quick actions and events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    pieces
    -1.02
    units
    -1.00
    stories
    -0.98
     constructs
    -0.97
    ometers
    -0.96
     pillars
    -0.95
    posts
    -0.93
    types
    -0.91
    killers
    -0.90
    ernels
    -0.90
    POSITIVE LOGITS
     invitation
    0.92
     copy
    0.87
     appointment
    0.87
     outing
    0.87
     trip
    0.86
     berth
    0.86
     cameo
    0.85
     apology
    0.85
     appearance
    0.85
     visit
    0.84
    Act Density 0.304%

    No Known Activations